Periodontitis, Smoking and Implant Survival: Changes in Hazard-Ratio over Time

Objectives: To estimate the effect of periodontal status upon hazard for dental implant failure using statistical methods that take into account both cluster observations among patients and the possibility that hazard ratios are not constant throughout long term follow up. Methods: A prospective cohort study of 736 patients with a total of 2336 dental implants was evaluated. Survival analysis methods were applied in order to estimate the cumulative survival rates. An extended Cox-Model with an interaction term between follow-up time and the variables was performed. Results: Cumulative survival analysis demonstrated that for healthy periodontal patients, the CSR stabilized around 60 months to a level of 0.96, while CSR continued to drop throughout follow up period for severe chronic periodontitis patients and reach a level of 0.88 at 108 month. The hazard ratios for both severe chronic periodontitis and smoking were not constant over time; both factors present higher hazard ratios at longer follow-up time. The final (extended) Cox model revealed that severe chronic periodontitis does not relate to a greater Hazard for implant failure up to 50 months, but turn out to be a strong significant risk factor after 50 months (H.R=8.06;p<0.01). The extended Cox model for smoking indicates a near significant hazard ratio of 2.76 (p=0.061) after 50 months. Conclusion: Severe chronic periodontal patients are at a major risk of implant failure after 50 months form placement. Hazard ratios for periodontal and smoking status are not constant throughout the follow-up period.